Guest  

 
Search: 
Search By: SubjectAbstractAuthorTitleFull-Text

 

Showing 1 through 5 of 443 records.
Pages: Previous - 1 2 3 4 5 6 7 8 9 10 11 12 13 ... 89 - Next  Jump:
2018 - ICA's 68th Annual Conference Words: 342 words || 
Info
1. Hearn, Alison. "Datafied Living and the ‘Citizen Score’: Credit Scoring, Soft Power and the Redefinition of Trust" Paper presented at the annual meeting of the ICA's 68th Annual Conference, Hilton Prague, Prague, Czech Republic, <Not Available>. 2019-11-19 <http://citation.allacademic.com/meta/p1364295_index.html>
Publication Type: Session Paper
Abstract: When credit-reporting agency Equifax was hacked in July of 2017, news coverage predictably criticized the failure of the company’s cyber-security mechanisms, which left consumers vulnerable to increased levels of identity theft. No one commented on the conditions that produced this kind of vulnerability – the sheer breadth and depth of consumer data collected by Equifax – or on the ways automated credit-scoring systems have long worked to shape and privilege specific kinds of social identity in the first place. The coverage of the Equifax hack not only demonstrates the degree to which our datafied lives have become unremarkable, but also the extent to which we are now required to trust such computational practices even when they fail, or risk economic and cultural exclusion. Ian Bogost calls this cultural condition ‘computational theocracy’; while automation and datafication are deemed the height of rationality, they simultaneously demand absolute faith from everyday users in their procedures (Bogost, 2015). This form of techno-fetishism obscures both the material conditions of production behind datafication and the ways in which it is displacing traditional truth arbiters and radically redefining the meaning of trust itself. Nowhere is this more evident than in China’s recent plan to establish a countrywide ‘social credit system’, which will assign each citizen a numerical score based on their computed level of trustworthiness and ‘sincerity’. Marketed as a way to ‘forge a public opinion environment where keeping trust is glorious’ (Planning Outline, 2014), the citizen score will consider five dimensions of individual conduct derived from online data, including personal behavior and the quality of social networks. This paper will consider some of the potential effects of living with the ‘citizen score’ via an examination of one the most prominent prototypes being considered for use by the Chinese government – Sesame Credit. Proprietary practices of big data-driven analytics may demand our trust to accumulate capital and entrench new forms of soft political power, but, given their efforts to contain and control our life chances with little accountability or transparency, there is certainly no reason we should oblige them.

2014 - AAAL Annual Conference Words: 46 words || 
Info
2. Jeffery, Jill. and Carhill-Poza, Avary. "Investigating the Representation of Adolescent L2 Writers in High-Stakes Assessment Scoring Materials: A Comparison of High- and Low-Scoring Writing Features" Paper presented at the annual meeting of the AAAL Annual Conference, Portland Marriott Downtown Waterfront, Portland, OR, Mar 22, 2014 <Not Available>. 2019-11-19 <http://citation.allacademic.com/meta/p698395_index.html>
Publication Type: Individual Paper
Review Method: Peer Reviewed
Abstract: This study investigates features of student writing that differentiate high and low scores in a corpus of US secondary “benchmark” writing assessment responses. The authors focus on the extent to which language features indicative of diverse linguistic repertoires might be implicated in the interpretation of results.

2007 - American Association of Colleges of Pharmacy Words: 265 words || 
Info
3. Layson-Wolf, Cherokee., Trovato, James., Petrelli, Heather. and Morgan, Jill. "A scoring tool to standardize evaluation of applicants for admissions and to evaluate scores as predictors of success" Paper presented at the annual meeting of the American Association of Colleges of Pharmacy, Disney’s Yacht & Beach Club Resort, Lake Buena Vista, Florida, Jul 14, 2007 <Not Available>. 2019-11-19 <http://citation.allacademic.com/meta/p196170_index.html>
Publication Type: Abstract
Abstract: Purpose: The purpose of this project is the development and utilization of a new admission application scoring tool to evaluate applicants applying to the University of Maryland, School of Pharmacy. This tool allows the admissions committee to compare candidates based on a group of factors as opposed to single factors such as PCAT or GPA. Methods: The admissions committee identified five target areas that are of major importance when considering a pharmacy school applicant for interview. These five target areas included in the tool are: letters of recommendation, academic performance, PCAT score, work experience, and evidence of leadership. A four point scale was created for each area creating the highest potential score of 20, with 1 as the lowest rating, and 4 as the highest rating. Each candidate is evaluated according to their application packet and the information is utilized in the admissions review process. After the interview process, scores are included based on interview results. Total scores are utilized in the admissions decision process. For those admitted and entering the school, we will compare their tool scores to measures such as GPA to evaluate any relationship between scoring and academic performance. Conclusions: At the end of this admissions cycle, we would have utilized this screening tool for two years and will continue to refine our admissions process. For the coming years, we will utilize the data to compare scores of accepted candidates and performance in the first year of the curriculum.

2011 - North American Chapter of the International Group for the Psychology of Mathematics Education Pages: unavailable || Words: unavailable || 
Info
4. Meylani, Rusen. and Teuscher, Dawn. "Using Neural-Networks to Predict AP-Calculus Test Scores from PCA and ACT Mathematics Test Scores" Paper presented at the annual meeting of the North American Chapter of the International Group for the Psychology of Mathematics Education, University of Nevada, Reno, Reno, NV, Oct 20, 2011 Online <APPLICATION/PDF>. 2019-11-19 <http://citation.allacademic.com/meta/p512666_index.html>
Publication Type: Poster
Review Method: Peer Reviewed
Abstract: Neural-Networks are a powerful alternative to regression especially for prediction and forecasting but not widely used in educational research. This study explores how AP-Calculus AB and BC scores can be predicted from the Precalculus Concept Assessment (PCA) and ACT mathematics scores employing two commonly used Neural-Networks models. Strong positive correlations between the actual and predicted values of the AP-Calculus exam scores confirm that Neural-Networks is an efficient tool for prediction. This can help identify the students who are at the risk of not passing the AP-Calculus exams and help students, parents and teachers take remedial measures in a timely manner.

2015 - SRCD Biennial Meeting Words: 462 words || 
Info
5. Rhodes, Katherine., Branum-Martin, Lee., Washington, Julie. and Patton-Terry, Nicole. "Scoring the Diagnostic Evaluation of Language Variation Screening Test (DELV-S): Investigating a Polytomous Scoring Scheme" Paper presented at the annual meeting of the SRCD Biennial Meeting, Pennsylvania Convention Center and the Philadelphia Marriott Downtown Hotel, Philadelphia, PA, Mar 19, 2015 <Not Available>. 2019-11-19 <http://citation.allacademic.com/meta/p961751_index.html>
Publication Type: Individual Poster
Review Method: Peer Reviewed
Abstract: The Diagnostic Evaluation of Language Variation screening test (DELV-S) is a popular screening instrument for distinguishing language variation from language disorder risk. Although the DELV-S features four categories for children's responses to language prompts, only two categories (variation from Mainstream American English and Mainstream American English consistent responses) are used in the scoring of the Language Variation Status subtest. Responses from children who vary from the desired response categories or who do not respond are not considered during scoring. Furthermore, the factor validity of the DELV-S has not been established (Petscher, Connor, Al Otaiba, 2012; Seymour, Roeper, & de Villiers, 2003), and previous studies have demonstrated that the Diagnostic Risk Status subtest does not display the single factor structure claimed by the test publishers (Petscher, Connor, Al Otaiba, 2012). Without an investigation of the scoring and scaling of the DELV-S, it is not possible to understand its factor structure, and without understanding the factor structure of the DELV-S it is difficult to interpret its construct validity.

This issue of construct validity is especially pertinent for African American English dialect (AAE) users, because these children have a history of over-identification with language disorder as a result of their language variation from Mainstream American English dialect (Washington & Craig, 1992). The purpose of the current study is to evaluate the polytomous scoring categories of the Language Variation Status subtest of this popular language screening instrument, as well as to examine the potential linguistic differences between children who select category A (variation from MAE), category B (MAE consistent language), category C (something else, something other than the intended response), and category D (no response). This investigation of scoring categories must necessarily coincide with an investigation of the factor structure of this instrument.

The current study will examine the factor structure of the DELV-S Language Variation Status subtest using confirmatory factor analysis with a sample of N=250 African American students in grades 1-4. Models will be considered using the recommended dichotomous scoring criteria as well as the polytomous (A-D) response categories recorded during administration. Each of the models will be estimated using robust weighted least squares (WLSMV) estimation in MPlus version 7 (Muthen & Muthen, 2012). Local fit statistics (e.g., indicator factor loadings, indicator residuals), global fit statistics (e.g., model χ2, RMSEA, CFI), and parsimony will be used to evaluate each of the hypothesized models.

Results from the current study of DELV-S factor validity will be discussed with a wide audience in mind, including practitioners, researchers, and psychometricians who are working with African American children who come from low-income homes. Special considerations for DELV-S score interpretation will be discussed with regard to children who have language differences from MAE dialect, children who have language disorders, and children who have special learning profiles.

Pages: Previous - 1 2 3 4 5 6 7 8 9 10 11 12 13 ... 89 - Next  Jump:

©2019 All Academic, Inc.   |   All Academic Privacy Policy